Modify btree to delete known-dead index entries without an actual VACUUM.

When we are about to split an index page to do an insertion, first look
to see if any entries marked LP_DELETE exist on the page, and if so remove
them to try to make enough space for the desired insert.  This should reduce
index bloat in heavily-updated tables, although of course you still need
VACUUM eventually to clean up the heap.

Junji Teramoto

6 files changed, 144 insertions, 16 deletions

diff --git a/src/backend/access/nbtree/README b/src/backend/access/nbtree/README
index a9466aa634a..e153a47a2a2 100644
--- a/src/backend/access/nbtree/README
+++ b/src/backend/access/nbtree/README
@@ -1,4 +1,4 @@
-$PostgreSQL: pgsql/src/backend/access/nbtree/README,v 1.12 2006/05/08 00:00:09 tgl Exp $
+$PostgreSQL: pgsql/src/backend/access/nbtree/README,v 1.13 2006/07/25 19:13:00 tgl Exp $
 
 This directory contains a correct implementation of Lehman and Yao's
 high-concurrency B-tree management algorithm (P. Lehman and S. Yao,
@@ -146,11 +146,12 @@ Because a page can be split even while someone holds a pin on it, it is
 possible that an indexscan will return items that are no longer stored on
 the page it has a pin on, but rather somewhere to the right of that page.
 To ensure that VACUUM can't prematurely remove such heap tuples, we require
-btbulkdelete to obtain super-exclusive lock on every leaf page in the index
-(even pages that don't contain any deletable tuples).  This guarantees that
+btbulkdelete to obtain super-exclusive lock on every leaf page in the index,
+even pages that don't contain any deletable tuples.  This guarantees that
 the btbulkdelete call cannot return while any indexscan is still holding
 a copy of a deleted index tuple.  Note that this requirement does not say
-that btbulkdelete must visit the pages in any particular order.
+that btbulkdelete must visit the pages in any particular order.  (See also
+on-the-fly deletion, below.) 
 
 There is no such interlocking for deletion of items in internal pages,
 since backends keep no lock nor pin on a page they have descended past.
@@ -320,6 +321,43 @@ positives, so long as it never gives a false negative.  This makes it
 possible to implement the test with a small counter value stored on each
 index page.
 
+On-the-fly deletion of index tuples
+-----------------------------------
+
+If a process visits a heap tuple and finds that it's dead and removable
+(ie, dead to all open transactions, not only that process), then we can
+return to the index and mark the corresponding index entry "known dead",
+allowing subsequent index scans to skip visiting the heap tuple.  The
+"known dead" marking uses the LP_DELETE bit in ItemIds.  This is currently
+only done in plain indexscans, not bitmap scans, because only plain scans
+visit the heap and index "in sync" and so there's not a convenient way
+to do it for bitmap scans.
+
+Once an index tuple has been marked LP_DELETE it can actually be removed
+from the index immediately; since index scans only stop "between" pages,
+no scan can lose its place from such a deletion.  We separate the steps
+because we allow LP_DELETE to be set with only a share lock (it's exactly
+like a hint bit for a heap tuple), but physically removing tuples requires
+exclusive lock.  In the current code we try to remove LP_DELETE tuples when
+we are otherwise faced with having to split a page to do an insertion (and
+hence have exclusive lock on it already).
+
+This leaves the index in a state where it has no entry for a dead tuple
+that still exists in the heap.  This is not a problem for the current
+implementation of VACUUM, but it could be a problem for anything that
+explicitly tries to find index entries for dead tuples.  (However, the
+same situation is created by REINDEX, since it doesn't enter dead
+tuples into the index.)
+
+It's sufficient to have an exclusive lock on the index page, not a
+super-exclusive lock, to do deletion of LP_DELETE items.  It might seem
+that this breaks the interlock between VACUUM and indexscans, but that is
+not so: as long as an indexscanning process has a pin on the page where
+the index item used to be, VACUUM cannot complete its btbulkdelete scan
+and so cannot remove the heap tuple.  This is another reason why
+btbulkdelete has to get super-exclusive lock on every leaf page, not only
+the ones where it actually sees items to delete.
+
 WAL considerations
 ------------------
 
diff --git a/src/backend/access/nbtree/nbtinsert.c b/src/backend/access/nbtree/nbtinsert.c
index 974ef92d187..597949aa2d9 100644
--- a/src/backend/access/nbtree/nbtinsert.c
+++ b/src/backend/access/nbtree/nbtinsert.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtinsert.c,v 1.141 2006/07/13 16:49:12 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtinsert.c,v 1.142 2006/07/25 19:13:00 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -65,6 +65,7 @@ static void _bt_pgaddtup(Relation rel, Page page,
 			 OffsetNumber itup_off, const char *where);
 static bool _bt_isequal(TupleDesc itupdesc, Page page, OffsetNumber offnum,
 			int keysz, ScanKey scankey);
+static void _bt_vacuum_one_page(Relation rel, Buffer buffer);
 
 
 /*
@@ -262,6 +263,8 @@ _bt_check_unique(Relation rel, IndexTuple itup, Relation heapRel,
 												 hbuffer) == HEAPTUPLE_DEAD)
 					{
 						curitemid->lp_flags |= LP_DELETE;
+						opaque->btpo_flags |= BTP_HAS_GARBAGE;
+						/* be sure to mark the proper buffer dirty... */
 						if (nbuf != InvalidBuffer)
 							SetBufferCommitInfoNeedsSave(nbuf);
 						else
@@ -424,11 +427,29 @@ _bt_insertonpg(Relation rel,
 		 */
 		bool		movedright = false;
 
-		while (PageGetFreeSpace(page) < itemsz &&
-			   !P_RIGHTMOST(lpageop) &&
-			   _bt_compare(rel, keysz, scankey, page, P_HIKEY) == 0 &&
-			   random() > (MAX_RANDOM_VALUE / 100))
+		while (PageGetFreeSpace(page) < itemsz)
 		{
+			Buffer		rbuf;
+
+			/*
+			 * before considering moving right, see if we can obtain enough
+			 * space by erasing LP_DELETE items
+			 */
+			if (P_ISLEAF(lpageop) && P_HAS_GARBAGE(lpageop))
+			{
+				_bt_vacuum_one_page(rel, buf);
+				if (PageGetFreeSpace(page) >= itemsz)
+					break;		/* OK, now we have enough space */
+			}
+
+			/*
+			 * nope, so check conditions (b) and (c) enumerated above
+			 */
+			if (P_RIGHTMOST(lpageop) ||
+				_bt_compare(rel, keysz, scankey, page, P_HIKEY) != 0 ||
+				random() <= (MAX_RANDOM_VALUE / 100))
+				break;
+
 			/*
 			 * step right to next non-dead page
 			 *
@@ -438,7 +459,7 @@ _bt_insertonpg(Relation rel,
 			 * pages won't do because we don't know when they will get
 			 * de-linked from the tree.
 			 */
-			Buffer		rbuf = InvalidBuffer;
+			rbuf = InvalidBuffer;
 
 			for (;;)
 			{
@@ -702,9 +723,9 @@ _bt_split(Relation rel, Buffer buf, OffsetNumber firstright,
 	ropaque = (BTPageOpaque) PageGetSpecialPointer(rightpage);
 
 	/* if we're splitting this page, it won't be the root when we're done */
-	/* also, clear the SPLIT_END flag in both pages */
+	/* also, clear the SPLIT_END and HAS_GARBAGE flags in both pages */
 	lopaque->btpo_flags = oopaque->btpo_flags;
-	lopaque->btpo_flags &= ~(BTP_ROOT | BTP_SPLIT_END);
+	lopaque->btpo_flags &= ~(BTP_ROOT | BTP_SPLIT_END | BTP_HAS_GARBAGE);
 	ropaque->btpo_flags = lopaque->btpo_flags;
 	lopaque->btpo_prev = oopaque->btpo_prev;
 	lopaque->btpo_next = BufferGetBlockNumber(rbuf);
@@ -1651,3 +1672,47 @@ _bt_isequal(TupleDesc itupdesc, Page page, OffsetNumber offnum,
 	/* if we get here, the keys are equal */
 	return true;
 }
+
+/*
+ * _bt_vacuum_one_page - vacuum just one index page.
+ *
+ * Try to remove LP_DELETE items from the given page.  The passed buffer
+ * must be exclusive-locked, but unlike a real VACUUM, we don't need a
+ * super-exclusive "cleanup" lock (see nbtree/README).
+ */
+static void
+_bt_vacuum_one_page(Relation rel, Buffer buffer)
+{
+	OffsetNumber	deletable[MaxOffsetNumber];
+	int				ndeletable = 0;
+	OffsetNumber	offnum,
+					minoff,
+					maxoff;
+	Page			page = BufferGetPage(buffer);
+	BTPageOpaque	opaque = (BTPageOpaque) PageGetSpecialPointer(page);
+
+	/*
+	 * Scan over all items to see which ones need deleted
+	 * according to LP_DELETE flags.
+	 */
+	minoff = P_FIRSTDATAKEY(opaque);
+	maxoff = PageGetMaxOffsetNumber(page);
+	for (offnum = minoff;
+		 offnum <= maxoff;
+		 offnum = OffsetNumberNext(offnum))
+	{
+		ItemId itemId = PageGetItemId(page, offnum);
+
+		if (ItemIdDeleted(itemId))
+			deletable[ndeletable++] = offnum;
+	}
+
+	if (ndeletable > 0)
+		_bt_delitems(rel, buffer, deletable, ndeletable);
+	/*
+	 * Note: if we didn't find any LP_DELETE items, then the page's
+	 * BTP_HAS_GARBAGE hint bit is falsely set.  We do not bother
+	 * expending a separate write to clear it, however.  We will clear
+	 * it when we split the page.
+	 */
+}
diff --git a/src/backend/access/nbtree/nbtpage.c b/src/backend/access/nbtree/nbtpage.c
index e069495f94f..080e10c88cf 100644
--- a/src/backend/access/nbtree/nbtpage.c
+++ b/src/backend/access/nbtree/nbtpage.c
@@ -9,7 +9,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtpage.c,v 1.98 2006/07/13 16:49:12 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtpage.c,v 1.99 2006/07/25 19:13:00 tgl Exp $
  *
  *	NOTES
  *	   Postgres btree pages look like ordinary relation pages.	The opaque
@@ -668,6 +668,15 @@ _bt_delitems(Relation rel, Buffer buf,
 	opaque = (BTPageOpaque) PageGetSpecialPointer(page);
 	opaque->btpo_cycleid = 0;
 
+	/*
+	 * Mark the page as not containing any LP_DELETE items.  This is not
+	 * certainly true (there might be some that have recently been marked,
+	 * but weren't included in our target-item list), but it will almost
+	 * always be true and it doesn't seem worth an additional page scan
+	 * to check it.  Remember that BTP_HAS_GARBAGE is only a hint anyway.
+	 */
+	opaque->btpo_flags &= ~BTP_HAS_GARBAGE;
+
 	MarkBufferDirty(buf);
 
 	/* XLOG stuff */
diff --git a/src/backend/access/nbtree/nbtutils.c b/src/backend/access/nbtree/nbtutils.c
index 840244e597a..8b680541121 100644
--- a/src/backend/access/nbtree/nbtutils.c
+++ b/src/backend/access/nbtree/nbtutils.c
@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtutils.c,v 1.77 2006/07/13 16:49:12 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtutils.c,v 1.78 2006/07/25 19:13:00 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -884,9 +884,15 @@ _bt_killitems(IndexScanDesc scan, bool haveLock)
 	 * Since this can be redone later if needed, it's treated the same
 	 * as a commit-hint-bit status update for heap tuples: we mark the
 	 * buffer dirty but don't make a WAL log entry.
+	 *
+	 * Whenever we mark anything LP_DELETEd, we also set the page's
+	 * BTP_HAS_GARBAGE flag, which is likewise just a hint.
 	 */
 	if (killedsomething)
+	{
+		opaque->btpo_flags |= BTP_HAS_GARBAGE;
 		SetBufferCommitInfoNeedsSave(so->currPos.buf);
+	}
 
 	if (!haveLock)
 		LockBuffer(so->currPos.buf, BUFFER_LOCK_UNLOCK);
diff --git a/src/backend/access/nbtree/nbtxlog.c b/src/backend/access/nbtree/nbtxlog.c
index 63cb342e393..126f593cd96 100644
--- a/src/backend/access/nbtree/nbtxlog.c
+++ b/src/backend/access/nbtree/nbtxlog.c
@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtxlog.c,v 1.35 2006/07/14 14:52:17 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/access/nbtree/nbtxlog.c,v 1.36 2006/07/25 19:13:00 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -346,6 +346,7 @@ btree_xlog_delete(XLogRecPtr lsn, XLogRecord *record)
 	Relation	reln;
 	Buffer		buffer;
 	Page		page;
+	BTPageOpaque opaque;
 
 	if (record->xl_info & XLR_BKP_BLOCK_1)
 		return;
@@ -374,6 +375,13 @@ btree_xlog_delete(XLogRecPtr lsn, XLogRecord *record)
 		PageIndexMultiDelete(page, unused, unend - unused);
 	}
 
+	/*
+	 * Mark the page as not containing any LP_DELETE items --- see comments
+	 * in _bt_delitems().
+	 */
+	opaque = (BTPageOpaque) PageGetSpecialPointer(page);
+	opaque->btpo_flags &= ~BTP_HAS_GARBAGE;
+
 	PageSetLSN(page, lsn);
 	PageSetTLI(page, ThisTimeLineID);
 	MarkBufferDirty(buffer);
diff --git a/src/include/access/nbtree.h b/src/include/access/nbtree.h
index b22422ffd6f..232cbb92bdb 100644
--- a/src/include/access/nbtree.h
+++ b/src/include/access/nbtree.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/nbtree.h,v 1.101 2006/07/11 21:05:57 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/access/nbtree.h,v 1.102 2006/07/25 19:13:00 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -71,6 +71,7 @@ typedef BTPageOpaqueData *BTPageOpaque;
 #define BTP_META		(1 << 3)	/* meta-page */
 #define BTP_HALF_DEAD	(1 << 4)	/* empty, but still in tree */
 #define BTP_SPLIT_END	(1 << 5)	/* rightmost page of split group */
+#define BTP_HAS_GARBAGE	(1 << 6)	/* page has LP_DELETEd tuples */
 
 
 /*
@@ -163,6 +164,7 @@ typedef struct BTMetaPageData
 #define P_ISROOT(opaque)		((opaque)->btpo_flags & BTP_ROOT)
 #define P_ISDELETED(opaque)		((opaque)->btpo_flags & BTP_DELETED)
 #define P_IGNORE(opaque)		((opaque)->btpo_flags & (BTP_DELETED|BTP_HALF_DEAD))
+#define P_HAS_GARBAGE(opaque)	((opaque)->btpo_flags & BTP_HAS_GARBAGE)
 
 /*
  *	Lehman and Yao's algorithm requires a ``high key'' on every non-rightmost